Skip to content

fix indexing and update pandas code per v3.0 changes #154

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
nkasica wants to merge 2 commits into
base: develop
Choose a base branch
from
Open

fix indexing and update pandas code per v3.0 changes #154

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
83509d6
fix indexing and update pandas per v3.0 changes
nkasica Feb 18, 2026
374d979
black formatting
nkasica Feb 19, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 0 additions & 1 deletion docs/conf.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,6 @@

import pkg_resources


# -- Project information -----------------------------------------------------

project = "pipit"
Expand Down
1 change: 0 additions & 1 deletion docs/examples/csv_reader.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,6 @@

import pipit as pp


if __name__ == "__main__":
# Use pipit's ``from_csv`` API to read in traces in CSV format.
# The result is stored into pipit's Trace data structure.
Expand Down
1 change: 0 additions & 1 deletion docs/examples/hpctoolkit.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,6 @@

import pipit as pp


if __name__ == "__main__":
# Path to HPCToolkit traces
dirname = "../../pipit/tests/data/ping-pong-hpctoolkit"
Expand Down
1 change: 0 additions & 1 deletion docs/examples/nsight.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,6 @@

import pipit as pp


if __name__ == "__main__":
# Path to Nsight traces
filename = "../../pipit/tests/data/nbody-nvtx/trace.csv"
Expand Down
1 change: 0 additions & 1 deletion docs/examples/otf2_read.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,6 @@

import pipit as pp


if __name__ == "__main__":
# Path to OTF2 traces
dirname = "../../pipit/tests/data/ping-pong-otf2"
Expand Down
1 change: 0 additions & 1 deletion docs/examples/projections.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,6 @@

import pipit as pp


if __name__ == "__main__":
# Path to OTF2 traces
dirname = "../../pipit/tests/data/ping-pong-projections"
Expand Down
21 changes: 13 additions & 8 deletions pipit/trace.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -277,14 +277,19 @@ def _match_caller_callee_by_level(filtered_df):
enter_leave_df = self.events.loc[enter_leave_mask]

# add dummy values for depth/parent/children
# (otherwise loc won't insert the values)
self.events["_depth"] = 0
self.events["_parent"] = None
self.events["_children"] = None
self.events.loc[enter_leave_mask] = enter_leave_df.groupby(
self.parallelism_levels, group_keys=False, dropna=False

dummy = enter_leave_df.groupby(
self.parallelism_levels, group_keys=False, dropna=False, observed=False
).apply(_match_caller_callee_by_level)

# ensure proper indexing alignment on insert
self.events.loc[enter_leave_mask, "_depth"] = dummy["_depth"]
self.events.loc[enter_leave_mask, "_parent"] = dummy["_parent"]
self.events.loc[enter_leave_mask, "_children"] = dummy["_children"]

self.events = self.events.astype({"_depth": "Int32", "_parent": "Int32"})
self.events = self.events.astype({"_depth": "category", "_parent": "category"})

Expand Down Expand Up @@ -638,7 +643,7 @@ def calc_idle_time(events):
return idle_time

return (
self.events.groupby(self.parallelism_levels, dropna=False)
self.events.groupby(self.parallelism_levels, dropna=False, observed=False)
.apply(
calc_idle_time,
)
Expand Down Expand Up @@ -683,7 +688,7 @@ def time_profile(self, num_bins=50, normalized=False):
self.calc_inc_metrics(["Timestamp (ns)"])

# Filter by Enter rows
events = self.events[self.events["Event Type"] == "Enter"].copy(deep=False)
events = self.events[self.events["Event Type"] == "Enter"]
names = events["Name"].unique().tolist()

# Create equal-sized bins
Expand All @@ -706,7 +711,7 @@ def calc_exc_time_in_bin(events):
}

# start out with exc times being a copy of inc times
exc_times = list(events["inc_time_in_bin"].copy(deep=False))
exc_times = list(events["inc_time_in_bin"])

# filter to events that have children
filtered_df = events.loc[events["_children"].notnull()]
Expand Down Expand Up @@ -741,7 +746,7 @@ def calc_exc_time_in_bin(events):
in_bin = events[
(events["_matching_timestamp"] > start)
& (events["Timestamp (ns)"] < end)
].copy(deep=False)
]

# Calculate inc_time_in_bin for each function
# Case 1 - Function starts in bin
Expand Down Expand Up @@ -776,7 +781,7 @@ def calc_exc_time_in_bin(events):
calc_exc_time_in_bin(in_bin)

# Sum across all processes
agg = in_bin.groupby("Name")["exc_time_in_bin"].sum()
agg = in_bin.groupby("Name", observed=False)["exc_time_in_bin"].sum()
profile.append(agg.to_dict())

# Convert to DataFrame
Expand Down